1. To generate new ES cell lines from rat embryos.[unreadable] [unreadable] Our generation of cell lines from the mouse epiblast (EpiSCs) is important for two reasons. First, this may be a step towards a general method to derive ES cells from any vertebrate. We will test this possibility by generating EpiSCs from the rat epiblast. Second, cells that represent the epiblast rather than the inner cells mass will permit better control of the differentiation of somatic cells. To date pluripotent embryonic stem cell lines have only been derived in a limited number of species including mouse, chicken, and primate. Rat ES cells have not been obtained previously. Rat ES cells would provide a major tool for neuroscience research. Preliminary data suggest rat ES cells can be derived from the post-implantation rat epiblast using conditions that support human ES and mouse EpiSCs. For transplantation studies involving behavioral analysis, the rat would be a better experimental animal. The production of genetically manipulated rats would extend the scope of molecular neuroscience to a mammal with different behaviors from the mouse. This goal depends on post-ES cells contributing to the germ line. Our current model suggests this will require a de-differentiation event. Our analysis of the differences between pre- and post-implantation ES cells will guide our choice of de-differentiation strategy. [unreadable] [unreadable] 2. To define the chromatin state of key pluripotency genes. [unreadable] [unreadable] There is great interest in how the stable epigenetic state of chromatin is maintained in ES cells. The derivation of EpiSCs cells allows a direct comparison of the state of chromatin in two distinct cells that are both pluripoitent. We propose to use chromatin immunoprecipitation (ChIP) methods to analyze the epigenetic state of a subset of genes that are known to maintain pluripotency and regulate early steps in gastrulation. [unreadable] [unreadable] Our preliminary analysis of the Oct4 regulatory region using ChIP shows that in post-ES cells, Oct4 itself is bound to the proximal enhancer, a region known in mice to regulate expression in the epiblast. ChIP data also shows that STAT3 binding in this region is increased in post-ES cells. The activin/nodal ALK receptors activate SMAD 2/3 and promote human ES cell self-renewal. In neural stem cells SMAD and STAT interact to promote astrocytic differentiation. We will determine the involvement of SMADs in the proximal enhancer of Oct4 to determine if STAT3 and SMADs are directly involved in maintaining the pluripotency of post-ES cells. These data illustrate how ChIP will provide information that is essential to understanding the link between survival signaling and epigenetic status. [unreadable] [unreadable] 3. To determine the heterogeneity of EpiSCs and human ES cells. [unreadable] [unreadable] The homogeneity of the cells is a central question in the ES cell field. We will approach this problem in EpiSCs cells first at the level of gene expression and then at the level of protein expression. ABI Inc. have developed a set of protocols and platforms to reproducibly study cells at the single-cell level. Two advances that have made this possible are the Pre-Amplification concept and the TaqMan Low Density Arrays. Our preliminary data suggests that all the cells express Nanog and Oct4 but none express Stella, a gene expressed in primitive germ cells. Genes associated with gastrulation are expressed more frequently in EpiSCs. These data suggest that single cell PCR will provide valuable information on ES cell homogeneity to use as a basis for subsequent studies using FACS analysis of protein expression. We have sub-cloned EpiSCs allowing us to determine if distinct types of epiblast cell can be maintained in culture. This information will set up a sound basis for further work on ES cell differentiation. [unreadable] [unreadable] 3. The differentiation of embryonic stem cells. [unreadable] [unreadable] Information on the homogeneity of ES cells is a pre-requisite for controlling their differentiation but preliminary work with human ES and mouse EpiSCs demonstrates that we have simple assays for the first steps in differentiation. The growth factor, BMP-2, is sufficient to induce brachyury and other genes indicative of mesendoderm induction. These data suggest that simple changes in the culture conditions differentiate either EpiSCs or human ES cells to mesendodermal fates. These fates are interesting to us because brachyury positive mesoderm includes the axial mesoderm that induces the floor plate fate and because we believe that an epiblast precursor chooses between a mesendodermal and ectodermal fate. [unreadable] [unreadable] 4.Defining an in vitro assay that predicts stem cell activation in vivo. [unreadable] [unreadable] The reason for our focus on the action of growth factors is that we have already shown that Notch ligands stimulate the stem cell survival in vitro and have powerful neurotrophic effects in vivo. Our main goal is to determine if in vitro analysis of the survival network predicts the in vivo neurotrophic effect of a stimulus. The imaging associated with this analysis will be automated using automatic microscopy and image processing. We are interested in developing simple assays that predict the in vivo effects on the stem cell niche. We can now analyze one thousand data points a week for a simple stem cell survival assay. We anticipate a very exciting set of fast assays allowing rapid optimization of regenerative therapies.[unreadable] [unreadable] 5. Lineages in CNS stem cells. [unreadable] [unreadable] Our imaging of CNS stem cells in culture generates lineages that definitively identify multipotent precursors for neurons and glia. About half the cells are tripotent stem cells and their decay kinetics shows them to be a uniform population. The acquisition of further data on their properties would be simplified if we had a surface marker to prospectively identify these tripotent cells. CD15 has been proposed to fill this need. The initial frequency of tripotential cells after passage matches the frequency of CD15-positive cells, suggesting that CD15 marks the tripotential stem cell. Further, the decay frequencies CD15 and tri-potent cells are also similar. We have also shown that CD15 recognizes a subset of cells in cultures acutely derived from human glioblastoma. We propose to analyze the relationship between CD15 expression and the stem cell state. [unreadable] [unreadable] Our lineage data has opened up a radically new view of fate choice by CNS stem cells. One of the novel aspects of our current results shows that fate choices occurs very rapidly and that CNTF triggers a subtle instructive change to generate astrocytes from a bi-potent cell. We have also shown that PDGF stimulates a rapid generation of oligodendrocyte precursors from CNS stem cells. We propose further experiments to determine if pro-neurogenic genes act in a similar instructive way on bi-potent precursors. Our goal is to define the fundamental fate choice mechanisms that control the generation of neurons, astrocytes and oligodendrocytes.[unreadable] [unreadable] 6. To determine neuronal survival mechanisms. [unreadable] [unreadable] To derive functional neurons from stem cells it is necessary to understand neuronal survival mechanisms. Using primary hippocampal neurons, we established that young neurons go through a restricted period when they die and neurotrophins are required for their survival. Pharmacological studies suggest that activation of the neurotrophin receptor is only transient and this sets up a stable activation of AKT mediated by L-type channels and integrins that controls survival by regulating the ratio of BclXL to Bax. Neuron death in the neurotrophin sensitive phase is dependent on p53. Neurons spontaneously leave the neurotrophin dependent state and their numbers stabilize. This work sets up a powerful in vitro assay to study survival signaling in neurons that complements our work in ES cells and CNS stem cells.